ON THE LAUWERIER FORMULATION OF THE

advertisement
ON THE LAUWERIER FORMULATION OF THE
TEMPERATURE FIELD PROBLEM IN OIL STRATA
LYUBOMIR BOYADJIEV, OGNIAN KAMENOV, AND SHYAM KALLA
Received 13 January 2005
Dedicated to Professor Dr. Ivan Dimovski, Institute of Mathematics and Informatics,
Bulgarian Academy of Sciences, Bulgaria, on the occasion of his 70th birthday
The paper is concerned with the fractional extension of the Lauwerier formulation of the
problem related to the temperature field description in a porous medium (sandstone)
saturated with oil (strata). The boundary value problem for the fractional heat equation
is solved by means of the Caputo differintegration operator D∗(α) of order 0 < α ≤ 1 and
the Laplace transform. The solution is obtained in an integral form, where the integrand
is expressed in terms of a convolution of two special functions of Wright type.
1. Introduction
The problem of describing the temperature field u = u(x, y,z,t) of the strata arises when
a hot water or steam is injected into the strata in order to facilitate the oil extraction
process.
Two cases of fluid injection are mainly considered: linear and radial injections. In the
linear case, a hot fluid is forced into the strata in the positive and negative x-directions
with constant velocity through an infinitely long vertical gallery. In the radial case, a hot
fluid is forced through an infinitely thin well, which is considered as a linear source of
incompressible fluid with positive volume rate.
The heat equation for a porous medium is derived in [1] under several generally accepted assumptions on the model. Beside the exact formulation of the temperature field
problem in oil strata, the following three approximate formulations are also treated:
(i) the lumped formulation, where the thermal conductivity of the strata is infinitely
large in the vertical direction;
(ii) the incomplete lumped formulation, where the horizontal heat transfer in the cap
and base rocks surrounding the strata is neglected;
(iii) the formulation of Lauwerier, where the horizontal heat transfer in the strata is
also neglected.
Our interest in this paper is particularly addressed to the fractional generalization of the
Lauwerier formulation since analogous problems concerning the lumped formulations
as well as the incomplete lumped formulation have been recently studied in [2]. The
Lauwerier formulation relates to the temperature field of a single layer stratum in the
case the velocity of the heat transfer between the fluid and the skeleton is finite. In this
Copyright © 2005 Hindawi Publishing Corporation
International Journal of Mathematics and Mathematical Sciences 2005:10 (2005) 1577–1588
DOI: 10.1155/IJMMS.2005.1577
1578
Fractional Lauwerier problem in oil strata
case, one has to consider separately the temperatures u(x,t) and Θ(x,z,t) of the fluid and
the cap rock, respectively. The heat equation for the regions containing the fluid and the
skeleton, respectively, is derived [1, Chapter 8, Section 8.7] under the main assumption
that instead of having two regions containing the fluid and the skeleton separately, there
is just a single region which is taken to be a porous medium. For a sufficiently large
filtration velocity, one can neglect the heat transferred to the cap rock and stratum in
the x-direction in comparison with the heat transfer in the z-direction. In this case, the
Lauwerier formulation for the linear fluid injection reads as
∂Θ ∂2 Θ
=
,
∂t
∂z2
0 < x,z,t < ∞,
z=0:
∂u
∂u
= −γ
− α(u − Θ),
∂t
∂x
0 < x,t < ∞,
z=0:
∂Θ
∂Θ
=µ
+ k(u − Θ),
∂t
∂z
0 < x,t < ∞,
(1.1)
x = 0 : u = 1,
2
(1.2a)
2
u,Θ −→ 0 as x + z −→ ∞,
(1.2b)
t = 0 : u = Θ = 0.
(1.2c)
The physical meaning of the constants involved can be briefly described as follows:
(i) the constant γ > 0 depends on the volume rate of the hot fluid forced into the
strata, the porosity specified as the ratio of the pores volume to the whole volume, and the coefficient of thermal conductivity of the cap rock;
(ii) the constant α > 0 depends on the porosity, the coefficients of thermal conductivity of the cap rock, and the volumic heat capacity of the fluid;
(iii) the constant µ > 0 depends on the coefficient of thermal conductivity of the cap
rock, the volumic heat capacity of the skeleton, and the coefficient of thermal
conductivity of the cap rock;
(iv) the constant k > 0 depends on the porosity and the volumic heat capacity of the
fluid and the skeleton.
By using the Laplace transform
f¯(p) = L f (t) =
∞
0
e− pt f (t)dt,
Re(p) > 0,
(1.3)
the solution of the problem (1.1) and (1.2) is given by the following formulas [1, Chapter
8, formulas (8.7.40) and (8.7.44)]:
u(x,t)
= e−αx/γ
t−x/γ
0



µ(2t − 2x/γ − τ) −(τµ/2√t−x/γ−τ)2
τµ

e−kτ  √
e
+ k erfc  3/2
2 π(t − x/γ − τ)
2 t − x/γ − τ
× I0 2
kαx
τ dτ,
γ
(1.4)
Lyubomir Boyadjiev et al.
Θ(x,z,t) = ke
−αx/γ

t−x/γ
e
0
−kτ
 z + µτ
kαx


erfc I0 2
dτ,
γτ
2 t − x/γ − τ
1579
(1.5)
where I0 (z) is the modified Bessel function of the first kind with zero index [6] and
2
erfc(x) = √
π
∞
x
e−t dt
2
(1.6)
is the complementary error function [5].
The main goal of this paper is to formulate in a reasonable way and to solve the fractional generalization of the problem (1.1) and (1.2).
Further on in the text we refer to this problem as fractional Lauwerier formulation of
the temperature field problem in oil strata.
2. Fractional heat equation
The concept of a noninteger differentiation and integration of a function is almost as
old as calculus itself. Many famous mathematicians including Leibniz, Euler, Lagrange,
Laplace, Fourier, and Abel made some contribution to it. Nowadays there exist specialized treaties where mathematical aspects and applications of the fractional calculus are
extensively discussed [13, 14, 17, 19].
Fractional calculus is a significant topic in mathematical analysis as a result of its increasing range of applications. One of the main advantages of the fractional calculus
is that the fractional derivatives provide an excellent instrument for the description of
memory and hereditary properties of various materials and processes. For our purpose,
we adopt in this paper the Caputo fractional derivative

t

f (m) (τ)
1



 Γ(m − β)
β+1−m dτ,
β
0 (t − τ)
D∗ f (t) = 
 d m f (t)



,
dt m
m − 1 < β < m,
(2.1)
β = m,
where m is a positive integer. This definition was introduced by Caputo in the late sixties
of the twentieth century and adopted by Caputo and Mainardi in the framework of the
theory of linear viscoelasticity. The method we apply makes the following rule of the
Laplace transform extremely important:
β
L D∗ f (t) = pβ f¯(p) −
m
−1
f k (0)pβ−1−k ,
m − 1 < β ≤ m.
(2.2)
k =0
The modelling of diffusion in a specific type of porous medium is one of the most significant applications of fractional-order derivatives [12, 18]. The fractional diffusion equation deduced by Nigmatullin [15, 16] in the form
∂2 u
∂2β u
= a2 2 ,
2β
∂t
∂z
1
0<β≤ ,
2
(2.3)
1580
Fractional Lauwerier problem in oil strata
deserves special attention. This equation is also know as the fractional diffusion-wave
equation [10, 11]. The reason is that when the order of the fractional derivative is 2β = 1,
the equation becomes the classical diffusion equation, and if 2β = 2, it becomes the classical wave equation.
For 0 < 2β < 1, we have the so-called ultraslow diffusion. Values 1 < 2β < 2 correspond
to the so-called intermediate processes [8].
In this paper, we consider the fractional diffusion equation (2.3) when the fractional
time derivative is determined by the Caputo fractional derivative (2.1). Thus we extend
the problem (1.1) and (1.2) to the fractional Lauwerier formulation of the temperature
field problem in oil strata that reads as
∂2 Θ
,
∂z2
2β
D∗ Θ =
1
0 < x,z,t < ∞, 0 < β ≤ ,
2
(2.4)
subject to the boundary conditions
∂u
− α(u − Θ), 0 < x,t < ∞,
∂x
∂Θ
2β
z = 0 : D∗ Θ = µ
+ k(u − Θ), 0 < x,t < ∞,
∂z
2β
z = 0 : D∗ u = −γ
(2.5)
and the conditions
x = 0,
z = 0 : u = 1,
2
(2.6a)
2
(2.6b)
t = 0 : u = Θ = 0.
(2.6c)
u,Θ −→ 0
as x + z −→ ∞,
3. Auxiliary results
With the idea to make this paper relatively self-contained, we summarize in this section
some important facts that enable us to solve the fractional problem stated above.
(i) First we mention that a key role in obtaining the solutions (1.4) and (1.5) is given
to the following theorem [3, pages 35–36].
Theorem 3.1 (Efros theorem). Assume there exist analytic functions G(p) and q(p) and
the relations
e−τ(p) G(p) = L g(t,τ) .
F(p) = L f (t) ,
(3.1)
Then
G(p)F q(p) = L
∞
0
f (τ)g(t,τ)dτ .
(3.2)
Theorem 3.1 is simply a generalization of the convolutional theorem for the Laplace
transform. Indeed, if we take q(p) = p, then
L g(t,τ) = e− pτ G(p),
(3.3)
Lyubomir Boyadjiev et al.
1581
and by the p-shift theorem, g(t,τ) = g(t − τ). Thus formula (3.9) becomes
G(p)F(p) = L
∞
0
f (τ)g(t,τ)dτ = L
t
0
f (t)g(t − τ)dτ ,
(3.4)
since for original functions we have g(t − τ) = 0 as t < τ.
(ii) The fundamental solution of the basic Cauchy problem for the time fractional
diffusion equation (2.3) can be expressed in terms of an auxiliary function defined as
[11]
M(z;β) =
1
2πi
Ha
eσ −zσ
dσ
,
σ 1−β
β
0 < β < 1,
(3.5)
where Ha denotes the Hankel path of integration that begins σ = −∞ − ib1 (b1 > 0), encircles the branch cut that lies along the negative real axis, and ends up at σ = −∞ + ib2
(b2 > 0). It is also proved that the following relation takes place:
M(z;β) = W(−z; −β;1 − β),
(3.6)
where
W(z;λ,µ) =
∞
zn
1
=
n!Γ(λn
+
µ)
2πi
n =0
Ha
eσ+zσ
−λ
dσ
,
dµ
λ > −1, µ > 0,
(3.7)
is an entire function of z referred to as the Wright function (cf. [7, Chapter 18]). In the
particular case β = 1/2, it holds that
M z;
m
∞
1
1 1
(−1)m
=√
2
π m=0
2
z2m
1
2/4
= √ e −z .
(2m)!
π
(3.8)
Further, we introduce the following auxiliary function:
N(ξ;β;λ) =
1
2πi
Ha
eσ −ξσ
2β
dσ
,
σλ
1
λ > 0, 0 < β ≤ ,
2
(3.9)
and prove the following statement.
Lemma 3.2. If 0 < β ≤ 1/2 and 0 < t,τ,z < ∞, then the following relations hold:
(i)
e−(z+µτ)p = L g1 (t,τ,z;β) ,
β
µ > 0;
(3.10)
(ii)
1 −(x/γ+τ)p2β
e
= L g2 (t,τ,β;λ) ,
λ
p
λ > 0, x > 0, γ > 0,
(3.11)
1582
Fractional Lauwerier problem in oil strata
where
g1 (t,τ,z,β) =
(z + µτ)β
z + µτ
M
;β ,
t β+1
tβ
g2 (t,τ;β;λ) = t λ−1 N
(3.12)
x/γ + τ
;β;λ .
t 2β
(3.13)
Proof. The validity of (i) is a direct consequence of [11, formulas (3.4) and (3.5)].
To prove part (ii), we consider the Laplace transform
1 −(x/γ+τ)p2β
e
,
pλ
L g2 (p,τ;β;λ) =
λ > 0.
(3.14)
According to the inversion formula for the Laplace transform,
g2 (t,τ;t;β;λ) =
1
2πi
Ha
e pt
1 −(x/γ+τ)p2β
e
d p.
pλ
(3.15)
dσ
.
σλ
(3.16)
Setting σ = pt and ξ = (x/γ + τ)/t 2β , we obtain
t λ −1
2πi
g2 (t,τ;β;λ) =
Ha
eσ −ξσ
2β
Hence, if 0 < 2β < 1,
g2 (t,τ;β;λ) = t λ−1 W(−ξ; −2β;λ) = t λ−1 N(ξ;β;λ).
(3.17)
In particular,
1
t λ −1
g2 t,τ; ;λ =
2
2πi
Ha
e
σ −ξσ
dσ
1
=
σ λ 2πi
Ha
e pt
1 −(x/γ+τ)p
e
d p.
pλ
(3.18)
Referring to [4, formulas (3) and (27)], we conclude that
1
g2 t,τ; ;λ =
2
t
0
s−λ−1
x
δ t − − τ − s ds,
Γ(λ)
γ
(3.19)
where Γ(λ) is the Euler gamma function and δ(t) is the Dirac delta function. It is also
worth mentioning that if λ = 1,
1
x
g2 t;τ; ,1 = H t − − τ ,
2
γ
where H(t − a) is the Heaviside function.
(3.20)
Lyubomir Boyadjiev et al.
1583
4. Fractional Lauwerier formulation
We apply the Laplace transform method to prove the following main results.
Theorem 4.1. If 0 < β ≤ 1/2 and λ > 0, the solutions of the fractional Lauwerier formulation
of the temperature field problem in oil strata (2.4), (2.5), and (2.6) are given by the integrals
u(x,t) =
Θ(x,z,t) = k
∞
0
∞
0
e
−(αx/γ+τk)
αkx
τ dτ,
γ
ϕ(t,τ;β) ∗ g1 (t,τ,0;β) I0 2
e−(αx/γ+τk) g1 (t,τ;z;β) ∗ g2 (t,τ,β;1) I0 2
(4.1)
αkx
τ dτ,
γ
(4.2)
where g1 (t,τ,z;β) and g2 (t,τ;β;1) are defined by (3.12) and (3.13), respectively, and
ϕ(t,τ;β) = g2 (t,τ;β;1 − 2β) + µg2 (t,τ;β;1 − β) + kg2 (t,τ;β;1).
(4.3)
Proof. We denote
ū(x, p) = L u(x,t) ,
Θ̄(x,z, p) = L Θ(x,z,t) ,
(4.4)
where L is the Laplace transformation operator. Applying the Laplace transform to (2.4),
(2.5), and (2.6), we obtain, in accordance with (2.2) and (2.6c),
p2β Θ̄(x,z, p) =
z = 0 : p2β ū(x, p) = −γ
z = 0 : p2β ū(x,z, p) = µ
∂2 Θ̄(x,z, p)
,
∂z2
0 < x, z < ∞,
∂ū(x, p)
− α ū(x, p) − Θ̄(x,z, p) ,
∂x
(4.5)
0 < x < ∞,
(4.6)
∂Θ̄(x,z, p)
+ k ū(x, p) − Θ̄(x,z, p) ,
∂x
1
x = 0 : ū(0, p) = ,
p
0 < x < ∞,
(4.7a)
ū,Θ −→ 0 as x2 + z2 −→ ∞.
(4.7b)
The solution of (4.5) which remains bounded as z → ∞ is
Θ̄(x,z, p) = c(x, p)e−zp ,
β
(4.8)
where c(x, p) is still unknown function. Substituting (4.8) into (4.6), we have
∂ū(x, p)
− α ū(x, p) − c(x, p) ,
∂x
(4.9)
p2β c(x, p) = −µpβ c(x, p) + k ū(x, p) − c(x, p) .
(4.10)
p2β ū(x, p) = −γ
Solving (4.10) for c(x, p), we get
c(x, p) =
kū(x, p)
.
p2β + µpβ + k
(4.11)
1584
Fractional Lauwerier problem in oil strata
The substitution of this representation of c(x, p) into (4.9) leads to the equation
∂ū(x, p)
1 2β
αk
=−
p + α − 2β
ū(x, p).
∂x
γ
p + µpβ + k
(4.12)
The solution of this equation that conforms to (4.7a) is
p2β + α x
1
αkx
1
· 2β
−
.
ū(x, p) = exp
β
p
γ
p + µp + k
γ
(4.13)
Then obviously
c(x, p) =
(p2β + α)x
k
1
αkx
1
· exp
· 2β
−
,
2β
β
β
p + µp + k p
γ
p + µp + k
γ
(4.14)
and from (4.8) it follows that
p2β + α x
k
αkx
1
e−zp
· 2β
−
.
exp
2β
β
β
p p + µp + k
γ
p + µp + k
γ
β
Θ(x,z, p) =
(4.15)
To apply Theorem 3.1, we represent
ū(x, p) = F q(p;β) G(x, p;β),
(4.16)
where
F q(p;β) =
1
αkx
1
·
,
exp
q(p;β)
γ q(p;β)
G(x, p;β) =
(4.17)
p2β + µpβ + k
x
exp − p2β + α .
p
γ
(4.18)
The well-known formula [4, page 247, (2.4.82)] enables us to get
F(p) = L f (t) = L I0 2
αkx
t
γ
.
(4.19)
Further, we consider the function
e−τq(p;β) G(x, p;β) =
p2β + µpβ + k
x
αx
exp −
+ τ p2β − τµpβ −
+ τk
p
γ
γ
.
(4.20)
From Lemma 3.2 and the convolution theorem, we obtain
1 −(x/γ+τ)p2β −τµpβ
e
e
= L g2 (t,τ;β;1 − 2β) ∗ g1 (t,τ,0;β) ,
p1−2β
1 −(x/γ+τ)p2β −τµpβ
e
e
= L g2 (t,τ;β;1 − β) ∗ g1 (t,τ,0;β) ,
1
−
β
p
1 −(x/γ+τ)p2β −τµpβ
e
= L g2 (t,τ;β;1) ∗ g1 (t,τ,0;β) .
e
p
(4.21)
Lyubomir Boyadjiev et al.
1585
Taking into account (4.20) as well as the formulas (4.21), we can write
e−τq(p;β) G(x, p;β) = L g(t,τ) ,
(4.22)
where
g(t,τ) = e−(αx/γ+τk) ϕ(t,τ;β) ∗ g1 (t,τ,0;β) .
(4.23)
The application of Theorem 3.1 leads directly to the solution (4.1).
Following the same idea, we represent Θ̄(x,z, p) as the product
Θ̄(x,z, p) = F q(p;β) G1 (x, p;β),
(4.24)
where F[q(p,β)] is defined by (4.17) and
G1 (x, p;β) =
k −zpβ −(x/γ)p2β −αx/γ
e
e
.
e
p
(4.25)
Obviously
e−τq(p;β) G1 (x, p;β) = ke−(αx/γ+kτ) e−(z+µτ)p
β
1 −(x/γ+τ)p2β
.
e
p
(4.26)
Then as a direct consequence of Lemma 3.2 and the convolution theorem, we obtain
e−τq(p;β) G1 (x, p;β) = ke−(αx/γ+kτ) g1 (t,τ,z,β) ∗ g2 (t,τ;β;1) .
(4.27)
Then (4.24), (4.27), and Theorem 3.1 lead to the solution (4.2) and this completes the
proof of the theorem.
Corollary 4.2. For the particular case β = 1/2, the solution (4.1) reduces to (1.4).
Proof. Because of (3.8), we evidently can write
√ 2
τµ
1
g1 t,τ,0;
= √ 3/2 e−(τµ/2 t) .
2
2 πt
(4.28)
According to (3.13) and [9, Section 17.13, formula (27)],
x/γ + τ 1
1
1
; ;0
g2 t,τ; ;0 = N
2
t
t
2
1
x
x
e pt e−(x/γ+τ)p d p = δ t − − τ H t − − τ .
=
2πi Ha
γ
γ
(4.29)
Likewise from (3.13), [9, Section 17.13, formula (3)], and the t-shifting theorem for the
Laplace transform it follows that
x/γ + τ 1 1
1 1
1
; ;
g2 t,τ; ;
= 1/2 N
2 2
t
t
2 2
H(t − x/γ − τ)
1
1
e pt 1/2 e−(x/γ+τ)p d p = √ .
=
2πi Ha
p
π (t − x/γ − τ)
(4.30)
1586
Fractional Lauwerier problem in oil strata
Taking into account (3.20), we obtain that


1
1
x
x
ϕ t,τ;
+ k H t − − τ .
= δ t − − τ + µ √ 2
y
γ
π (t − x/γ − τ)
(4.31)
Formulas (4.28) and (4.31) yield that in the case β = 1/2 the convolution into the integrand of (4.1) takes the form
τµH(t − x/γ − τ)
1
√
∗ g1 (t,τ,0;β) =
ϕ t,τ;
2
2 π
+
t−x/γ−τ
1
s3/2
0
τµ2 H(t − x/γ − τ)
2π
kτµ
x
+ √ H t− −τ
2 π
γ
t
x/γ+τ
e
√
−(τµ/2 s)2
x
δ t − − τ − s ds
γ
√
1
1
2
e−(τµ/2 t−s) ds
3/2
(t − s)
(s − x/γ − τ)
t
x/γ+τ
√
1
−(τµ/2 t −s)2
e
ds.
(t − s)3/2
(4.32)
The third integral in (4.32) can be computed as
τµ
2
t
x/γ+τ
√
3/2 −(tµ/2 t −s)
(t − s) e
2
ds = 2
=2
t−x/γ−τ
e
0
∞
√
(τµ/2

√
t −x/γ−τ −s )2
−(τµ/2
t −x/γ−τ)
e
−w 2

τµ

d 2 t − x/γ − τ − s

√

τµ
.
dw = π erfc  2 t − x/γ − τ
(4.33)
By using [9, Section 3.471, formula (3)], one can see that
t
x/γ+τ
(t − s)
x
s− −τ
γ
3/2
−1/2
e
√
−(tµ/2 t −s)2
ds =
t−x/γ−τ
0
x
−3/2
x
t − −τ −x
γ
−1/2
√
e−(τµ/2
x)2
dx
√
=
√
2
π
2
e−(τµ/2 t−x/γ−τ) .
τµ (t − x/γ − τ)
(4.34)
For the first integral in (4.32), it is clear that
t−x/γ−τ
0
1
√
e(τµ/2
s3/2
s)2
δ t−
x
x
− τ − s ds = t − − τ
γ
γ
−3/2
√
e−(τµ/2
t −x/γ−τ)2
.
(4.35)
Lyubomir Boyadjiev et al.
1587
The substitution of these integrals by their expressions into (4.32) results in
ϕ t,τ;
1
∗ g1 (t;τ,0;β)
2





 µ2(t − x/γ) − τ 
√
τµ
2
−(τµ/2 t −x/γ−τ)
  ·H t− x −τ ;
√
=
e
+
k
erfc

γ
 2 π(t − x/γ − τ)3/2

2 t − x/γ − τ 
(4.36)
that proves the statement.
Corollary 4.3. For the particular case β = 1/2, the solution (4.2) reduces to (1.5).
Proof. From (3.8) and (3.12), it follows that
z + µτ 1 −((z+µτ)/2√t)2
1
·√ e
.
=
2
2t 3/2
π
g1 t,τ,z;
(4.37)
Then according to (3.20), the convolution into the integrand of (4.2) takes the form
g1 t,τ,z;
=
t
0
1
1
∗ g2 t,τ; ;1
2
2
√
z + µτ
x
2
e−((z+µτ)/2 t−s) H s − − τ ds
3/2
2 π(t − s)
γ
√
t
x
1
= H t− −τ √
γ
=H t−
2 π
x
1
−τ √
γ
2 π
x/γ+τ
z + µτ −((z+µτ)/2√t−s)2
e
ds
(t − s)3/2
t−x/γ−τ
√
z + µτ
2
e−((z+µτ)/2 t−x/γ−τ −ξ) dξ
3/2
(t − x/γ − τ − ξ)
0

x
2 t−x/γ−τ −((z+µτ)/2√t−x/γ−τ −ξ)2 
= H t− −τ √
e
d γ
π
γ
π
√
(z+µτ)/2

z + µτ
2 t − x/γ − τ − ξ
0
x
2 ∞
= H t− −τ √

t −x/γ−τ
(4.38)

e−w dw
2

z + µτ
x
,
= H t − − τ erfc  γ
2 t − x/γ − τ
and proves the statement.
Acknowledgment
The first author was partially supported by NSF-Bulgarian Ministry of Education and
Science under Grant no. MM 1305/2003.
1588
Fractional Lauwerier problem in oil strata
References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
M. Ya. Antimirov, A. A. Kolyshkin, and R. Vaillancourt, Applied Integral Transforms, CRM
Monograph Series, vol. 2, American Mathematical Society, Rhode Island, 1993.
L. Boyadjiev and R. Scherer, Fractional extensions of the temperature field problem in oil strata,
Kuwait J. Sci. Engrg. 31 (2004), no. 2, 15–32.
V. A. Ditkin and A. P. Prudnikov, Integral Transforms and Operational Calculus, Pergamon
Press, Oxford, 1965.
, Formulaire pour le calcul opérationnel, Masson & Cie, Paris, 1967.
A. Erdélyi, W. Magnus, F. Oberhettinger, and F. G. Tricomi, Higher Transcendental Functions.
Vol. I, McGraw-Hill, New York, 1953.
, Higher Transcendental Functions. Vol. II, McGraw-Hill, New York, 1953.
, Higher Transcendental Functions. Vol. III, McGraw-Hill, New York, 1955.
R. Gorenflo and R. Rutman, On ultraslow and on intermediate processes, Transform Methods
and Special Functions (P. Rusev, I. Dimovski, and V. Kiryakova, eds.), SCT Publishers, Singapore, 1995, pp. 61–81.
I. S. Gradshteyn and I. M. Ryzhik, Table of Integrals, Series, and Products, Academic Press, New
York, 1980.
F. Mainardi, On the initial value problem for the fractional diffusion-wave equation, Waves and
Stability in Continuous Media (Bologna, 1993) (S. Rionero and T. Ruggeri, eds.), Ser. Adv.
Math. Appl. Sci., vol. 23, World Scientific, New Jersey, 1994, pp. 246–251.
, Fractional relaxation-oscillation and fractional diffusion-wave phenomena, Chaos Solitons Fractals 7 (1996), no. 9, 1461–1477.
B. B. Mandelbrot, The Fractal Geometry of Nature, W. H. Freeman and Company, California,
1982.
A. C. McBride, Fractional Calculus and Integral Transforms of Generalized Functions, Research
Notes in Mathematics, vol. 31, Pitman, Massachusetts, 1979.
K. S. Miller and B. Ross, An Introduction to the Fractional Calculus and Fractional Differential
Equations, A Wiley-Interscience Publication, John Wiley & Sons, New York, 1993.
R. R. Nigmatullin, To the theoretical explanation of the “universal response”, Phys. Status Solidi
(B) 123 (1984), 739–745.
, The realization of the generalized transfer equation in a medium with fractal geometry,
Phys. Status Solidi (B) 133 (1986), 425–430.
K. B. Oldham and J. Spanier, The Fractional Calculus, Mathematics in Science and Engineering,
vol. 111, Academic Press, New York, 1974.
I. Podlubny, Fractional Differential Equations, Mathematics in Science and Engineering, vol.
198, Academic Press, California, 1999.
S. G. Samko, A. A. Kilbas, and O. I. Marichev, Fractional Integrals and Derivatives. Theory and
Applications, Gordon and Breach Science, Yverdon, 1993.
Lyubomir Boyadjiev: Institut für Praktische Mathematik, Universität Karlsruhe (TH), 76128 Karlsruhe, Germany
E-mail address: lyubomir [email protected]
Ognian Kamenov: Department of Applied Mathematics and Informatics, Technical University of
Sofia, P.O. Box 384, 1000 Sofia, Bulgaria
E-mail address: [email protected]
Shyam Kalla: Department of Mathematics and Computer Science, Kuwait University, P.O. Box
5969, Safat 13060, Kuwait
E-mail address: [email protected]
Download